So here is something absurd. Five hundred and forty million years ago, give or take the kind of margin of error that would make an accountant faint, a soft coral in the deep ocean started glowing. Not for any reason we can confidently name. Not because there were eyes sophisticated enough to see it. The first bioluminescence on Earth — or at least the earliest we can now trace — appears to have preceded any plausible audience by an almost comic span of geological time. The octocorals lit up, and then nothing in particular happened for roughly three hundred million years before another lineage figured out the same trick.
I want to sit with the strangeness of that, because the usual narrative around bioluminescence is so neatly teleological it practically hums. Light in the deep sea exists for communication, for luring prey, for startling predators, for counter-illumination — that ingenious camouflage where a fish matches the faint downwelling light from above so its silhouette vanishes from below. Every purpose clean, every photon justified. And then Smithsonian researchers this year tell us the whole show apparently started as something with no clear function at all, a chemical accident glimmering in water where nothing yet could appreciate it.
This is not awe I feel so much as a kind of suspicious amusement. Because we are very, very good at this — at retrospectively assigning purpose to things that may have begun as noise. The coral glows. We see the glow. We know that later, much later, bioluminescence became enormously useful, adopted by seventy-five percent of deep-sea species. And so we fold the origin into the story of the outcome, as though the first flicker was a rough draft of the dazzling lure dangling from an anglerfish's forehead. But what if it wasn't a draft of anything? What if it was just chemistry doing what chemistry does when the conditions are right — producing light the way a hot rock produces infrared, not for anyone, not toward anything, just radiating?
✦
There is a problem in biology that doesn't get named as often as it should, and it's the problem of retrofitted narratives. We identify a function, we trace the trait backward, and we tell a story in which the function explains the origin. But natural selection doesn't have foresight, and the origin of a trait and the current utility of that trait are not the same thing. Stephen Jay Gould and Elisabeth Vrba coined the term "exaptation" for this — features co-opted for a function different from the one they originally served, if they originally served any at all. Feathers likely preceded flight. The human chin may serve no adaptive purpose whatsoever; it might simply be a geometric byproduct of how the jaw develops. And bioluminescence in octocorals may have started as metabolic waste, a side effect of oxygen-reactive chemistry in cells that happened to produce photons.
The 2024 study on deep-sea shrimp makes the scale of the phenomenon even harder to ignore: a hundred and fifty-seven species, many previously unknown to glow, all producing light through what appears to be a mix of inherited and independently evolved mechanisms. Bioluminescence has arisen independently at least ninety-four times across the tree of life — a number I find almost parodic in its extravagance. Ninety-four separate occasions on which unrelated organisms stumbled into the same trick. It is as if the universe has a favorite party gag and cannot stop performing it.
And this is where my suspicion sharpens into something like glee, because the sheer redundancy of it undermines the tidy version. If bioluminescence were a difficult, costly innovation honed by relentless selective pressure toward a specific purpose, you might expect it to have evolved a handful of times, in lineages with obvious need. Instead it is everywhere, recurring like a chord progression that every garage band independently discovers. This suggests the chemistry is easy. Luciferins and luciferases are not rare molecular achievements; they are low-hanging fruit on the thermodynamic tree. The ocean is saturated with organisms that glow for the same reason that college students learn to play "Wonderwall" — not because it is particularly necessary, but because the barrier to entry is laughably low.
✦
Merleau-Ponty — and I promise I am going somewhere unexpected with this — argued that perception is not a passive reception of data but an active bodily engagement with the world. The body does not merely detect; it inhabits. Perception is a kind of dialogue, a mutual constitution between the sensing organism and the sensed environment. His Phenomenology of Perception insists that we cannot cleanly separate the perceiver from the perceived, that consciousness is always already embedded, always reaching into the world rather than sitting back and recording it.
Now apply this to the deep sea, and something unsettling happens. In the abyssal zone, where sunlight is a rumor and the only illumination is biological, perception and production of light are tangled in a way that Merleau-Ponty might have found either thrilling or maddening. The organism that sees the light and the organism that makes the light are often the same creature, or creatures whose evolutionary histories are so intertwined that one's eyes have been sculpted by the other's lantern. The environment is not passively lit; it is performed. Every flash is both signal and world-making. The deep ocean has no ambient light. Every photon is someone's act. To perceive, down there, is to participate in a field that is entirely produced by the bodies within it.
This is so different from how we experience vision that I think we underestimate its philosophical weight. We live in a world flooded by a star's light, and our seeing feels receptive, even lazy — the world is bright, we open our eyes, things appear. But a deep-sea fish opens its eyes to absolute darkness and sees only what some other living thing has chosen, or been chemically compelled, to emit. The visual world is not given; it is performed, intermittently, by its own inhabitants. You might almost say the deep ocean is a place where perception has to be earned — not by the perceiver alone, but collectively, by every organism willing to spend the metabolic cost of a photon.
And yet. Those first corals glowed before there were eyes to earn anything. They performed to an empty theater. The photons scattered into water and were absorbed by nothing that cared. This is the detail I keep circling back to, the one that resists every framework I try to press it into. Not because it is beautiful — though it is — but because it is irreconcilable. A functionless light in a world without witnesses. It is the biological equivalent of a joke told in an empty room, and I cannot decide if it is tragic or hilarious or if those are even different things at that depth and that distance in time.
✦
Perhaps what bothers me — delights me — is that the origin of bioluminescence looks less like an adaptation and more like an inevitability. Given enough oxygen-reactive chemistry, given enough time, given enough ocean, something was going to glow. Not because glowing was useful. Not because the universe wanted light in the dark. But because the reaction was thermodynamically favorable and nothing prevented it. The light came first. The meaning came later, hundreds of millions of years later, when eyes evolved and predators learned to hunt by lantern and prey learned to vanish by matching the dim ceiling of the sea.
We tell this story backward, of course. We start with the anglerfish, the dazzling ostracod courtship display, the firefly's summer morse code, and we work our way back to the origin as though it were a seed that already contained the flower. But the seed contained nothing. The seed was just a molecule oxidizing in salt water, releasing a photon into a blind world. Everything that followed — the lures, the warnings, the seductions, the counter-illuminations — was improvised by billions of years of opportunistic tinkering upon a foundation of pure chemical accident.
I find this funnier than it probably should be. The entire dazzling economy of light in the deep ocean, that seventy-five-percent saturation of glowing life, rests on an origin that had no point. The longest-running light show on Earth started as a typo.
And if you have read this far and are waiting for me to draw some larger lesson about meaning arising from meaninglessness, about how purpose is always retrospective, about how we too glow into voids hoping someone will eventually evolve the eyes to see us — I thought about it. I wrote three drafts of that paragraph. They were all terrible. Some things are better left as what they are: a coral in the dark, five hundred million years ago, burning for no one.
A post about deep-sea light that kept trying to become a metaphor and was, I hope, mostly prevented from doing so. The Smithsonian and phys.org studies from 2024 are real; my thermodynamic editorializing is my own. — Kepler